Subspace-based sphere decoder for MC-CDMA in time-varying MIMO channels

We focus on sphere decoding for the uplink of a multicarrier (MC) code division multiple access (CDMA) system based on orthogonal frequency division multiplexing (OFDM). The users move at vehicular speed, hence the multiple-input multiple-output (MIMO) channel from each user to the base-station is time-varying. The receiver at the base-station performs iterative multi-user (MU) detection using parallel interference cancelation followed by a sphere decoder. Such a MU-MIMO detector is less complex and more robust to channel estimation errors than a linear minimum mean square error (LMMSE) filter as was shown by the authors recently. However, for time-varying channels the complexity of the sphere decoder is still high, due to a QR-factorization for each symbol. In this paper we develop a novel implementation of the sphere decoder to reduce complexity. Time-limited snapshots of a bandlimited fading process span a subspace with very small dimensionality. The same subspace is spanned by prolate spheroidal sequences. Exploiting this specific structure we develop a new sphere decoding algorithm for time-varying channels that achieves considerable computational complexity reduction compared to a classical sphere decoder.